Shared-Memory Parallelism and Low-Rank Approximation Techniques Applied to Direct Solvers in FEM Simulation

Author

Amestoy, Patrick ; Buttari, Alfredo ; Joslin, Guillaume ; L´Excellent, Jean-Yves ; Sid-Lakhdar, Mohamed ; Weisbecker, Clement ; Forzan, Michele ; Pozza, Cristian ; Perrin, Remy ; Pellissier, Valene

Author_Institution

Inst. de Rech. en Inf. de Toulouse, Inst. Nat. Polytech. de Toulouse, Toulouse, France

Volume

50

Issue

2

fYear

2014

fDate

Feb. 2014

Firstpage

517

Lastpage

520

Abstract

In this paper, the performance of a parallel sparse direct solver on a shared-memory multicore system is presented. Large size test matrices arising from finite element simulation of induction heating industrial applications are used to evaluate the performance improvements due to low-rank representations and multicore parallelization.

Keywords

approximation theory; electrical engineering computing; finite element analysis; induction heating; mathematics computing; matrix algebra; multiprocessing systems; parallel processing; shared memory systems; FEM simulation; induction heating; induction heating industrial applications; large size test matrices; low-rank approximation techniques; low-rank representations; multicore parallelization; parallel sparse direct solver; shared-memory multicore system; shared-memory parallelism; Approximation methods; Eddy currents; Electromagnetic heating; Linear systems; Memory management; Parallel processing; Sparse matrices; Eddy currents; FEMs; linear systems; parallel algorithms; sparse matrices;

fLanguage

English

Journal_Title

Magnetics, IEEE Transactions on

Publisher

ieee

ISSN

0018-9464

Type

jour

DOI

10.1109/TMAG.2013.2284024

Filename

6748995